Flag Commit: Supporting Efficient Transaction Recovery in Flash-Based DBMSs

On, Sai Tung; Xu, Jianliang; Choi, Byron; Hu, Haibo; He, Bingsheng

doi:10.1109/tkde.2011.122

Cited by 19 publications

(16 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…FlashLogging [5] uses multiple USB flash sticks to boost logging performance. Flag Commit [31] embeds transaction status into flash pages by a chain of commit flags to minimize the need of logging. Although these proposals differ in how they achieve improved logging performance, they all remain thoroughly centralized and thus subject to the kinds of contention we propose to eliminate.…”

Section: Related Workmentioning

confidence: 99%

Scalable logging through emerging non-volatile memory

2014

View full text Add to dashboard Cite

Emerging byte-addressable, non-volatile memory (NVM) is fundamentally changing the design principle of transaction logging. It potentially invalidates the need for flush-beforecommit as log records are persistent immediately upon write. Distributed logging-a once prohibitive technique for single node systems in the DRAM era-becomes a promising solution to easing the logging bottleneck because of the nonvolatility and high performance of NVM.In this paper, we advocate NVM and distributed logging on multicore and multi-socket hardware. We identify the challenges brought by distributed logging and discuss solutions. To protect committed work in NVM-based systems, we propose passive group commit, a lightweight, practical approach that leverages existing hardware and group commit. We expect that durable processor cache is the ultimate solution to protecting committed work and building reliable, scalable NVM-based systems in general. We evaluate distributed logging with logging-intensive workloads and show that distributed logging can achieve as much as ∼3x speedup over centralized logging in a modern DBMS and that passive group commit only induces minuscule overhead.

show abstract

Section: Related Workmentioning

confidence: 99%

Scalable logging through emerging non-volatile memory

2014

View full text Add to dashboard Cite

show abstract

“…Simultaneously, the transaction abort ratio was set to 5% by default. As shown in [14], the cost of partial programming was also same as that of a page write. So we summarize the default parameter settings in table 1.…”

Section: Experiments Setupmentioning

confidence: 93%

“…For flash-based DBMSs built on SLC flash disks, based on shadow paging, Sai Tung On et al [14] proposed a new flagcommit scheme for efficient transaction recovery. It addresses these issues presented in [12].…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

MixSL: An Efficient Transaction Recovery Model in Flash-Based DBMS

Fan

Meng

2013

Web-Age Information Management

View full text Add to dashboard Cite

Abstract. With the development of flash technologies, flash disks have become an alternative to hard disk as external storage media. Because of the unique characteristics of flash disks such as fast random read access and out-place update, shadow paging technology can be adopted to support transaction recovery in flash-based DBMS. Inspired by shadow paging and logging, we propose a new transaction commit model named MixSL which can be used in databases built on MLC flash disks. Based on MixSL, we detail normal processing, garbage collection and recovery. For improving system performance and raising the utilization ratio of flash disks, we extend MixSL to support group commit. Our performance evaluation based on the TPC-C benchmark shows that MixSL outperforms the state-of-the-art recovery protocols.

show abstract

“…There has been significant recent research on the architectural evolution of the storage system with flash memory, including interface extensions for intelligent flash management [15], [16], [17], [19], [22] and system optimizations to exploit the flash memory advantages [6], [7], [12], [13]. In this section, we mainly focus on transaction support with flash memory.…”

Section: Related Workmentioning

confidence: 99%

LightTx: A lightweight transactional design in flash-based SSDs to support flexible transactions

Shu

Guo

et al. 2013

2013 IEEE 31st International Conference on Computer Design (ICCD)

View full text Add to dashboard Cite

Abstract-Flash memory has accelerated the architectural evolution of storage systems with its unique characteristics compared to magnetic disks. The no-overwrite property of flash memory has been leveraged to efficiently support transactions, a commonly used mechanism in systems to provide consistency. However, existing transaction designs embedded in flash-based Solid State Drives (SSDs) have limited support for transaction flexibility, i.e., support for different isolation levels between transactions, which is essential to enable different systems to make tradeoffs between performance and consistency. Since they provide support for only strict isolation between transactions, existing designs lead to a reduced number of on-the-fly requests and therefore cannot exploit the abundant internal parallelism of an SSD. There are two design challenges that need to be overcome to support flexible transactions: (1) enabling a transaction commit protocol that supports parallel execution of transactions; and (2) efficiently tracking the state of transactions that have pages scattered over different locations due to parallel allocation of pages.In this paper, we propose LightTx to address these two challenges. LightTx supports transaction flexibility using a lightweight embedded transaction design. The design of LightTx is based on two key techniques. First, LightTx uses a commit protocol that determines the transaction state solely inside each transaction (as opposed to having dependencies between transactions that complicate state tracking) in order to support parallel transaction execution. Second, LightTx periodically retires the dead transactions to reduce transaction state tracking cost. Experiments show that LightTx provides up to 20.6% performance improvement due to transaction flexibility. LightTx also achieves nearly the lowest overhead in garbage collection and mapping persistence compared to existing embedded transaction designs.

show abstract

Flag Commit: Supporting Efficient Transaction Recovery in Flash-Based DBMSs

Cited by 19 publications

References 22 publications

Scalable logging through emerging non-volatile memory

Scalable logging through emerging non-volatile memory

MixSL: An Efficient Transaction Recovery Model in Flash-Based DBMS

LightTx: A lightweight transactional design in flash-based SSDs to support flexible transactions

Contact Info

Product

Resources

About